field notes from vibe-driven developmenthttps://agentical.blog/en-ushttps://agentical.blog/blog/2026-05-24-cursor-4-hour-run/https://agentical.blog/blog/2026-05-24-cursor-4-hour-run/1,200 lines of code, three dead-ends, one saved afternoon. but only after i rewrote the prompt. a real debrief on 4 hours of unattended cursor agent.Sun, 24 May 2026 00:00:00 GMTfield-notescursoragentsworkflowsanna chenhttps://agentical.blog/blog/2026-05-21-cursor-to-claude-pipeline/https://agentical.blog/blog/2026-05-21-cursor-to-claude-pipeline/use cursor for the chunky edits, claude code for the long-running tasks. one pattern, three weeks of trial, two specific rules that made it stick.Thu, 21 May 2026 00:00:00 GMTpatternscursorclaude-codepipelinemarco silvahttps://agentical.blog/blog/2026-05-18-vibecheck-claude-code/https://agentical.blog/blog/2026-05-18-vibecheck-claude-code/a week of timing every commit. the answer is: yes, but only for a specific kind of work — and the gap is smaller than the discourse implies.Mon, 18 May 2026 00:00:00 GMTtoolingclaude-codebenchmarkspriya ramanhttps://agentical.blog/blog/2026-05-14-agents-md-guide/https://agentical.blog/blog/2026-05-14-agents-md-guide/AGENTS.md is the file every coding agent reads before doing anything. mine is 90 lines. here's the structure that's survived three months of daily use.Thu, 14 May 2026 00:00:00 GMTworkflowsagents-mdworkflowsconventionsjamie okonkwohttps://agentical.blog/blog/2026-05-10-checklist-pattern/https://agentical.blog/blog/2026-05-10-checklist-pattern/a pattern for writing prompts that survive long agentic runs. numbered steps, explicit verification, a single 'done' criterion. nothing else works as well.Sun, 10 May 2026 00:00:00 GMTpatternspromptingpatternsagentsanna chen